A new metric assesses the progress of AI models.Credit: Adapted from Jonathan Raa/NurPhoto/Getty
Today’s artificial intelligence (AI) systems can’t beat humans on long tasks, but they’re improving at a rapid pace and could close the gap sooner than many anticipated, according to an analysis of leading models1.
METR, a non-profit organization in Berkeley, California, created nearly 170 real-world tasks in coding, cybersecurity, general reasoning and machine learning, then established a…
Article Source
https://www.nature.com/articles/d41586-025-00831-8
